Genome-wide association study of eigenvectors provides genetic insights into selective breeding for tomato metabolites

Abstract

Background Long-term domestication and intensive breeding of crop plants aim to establish traits desirable for human needs, and characteristics related to yield, disease resistance, and postharvest storage have traditionally received considerable attention. These processes have led also to negative consequences, as is the case of loss of variants controlling fruit quality, for instance in tomato. Tomato fruit quality is directly associated to metabolite content profiles; however, a full understanding of the genetics affecting metabolite content during tomato domestication and improvement has not been reached due to limitations of the single detection methods previously employed. Here, we aim to reach a broad understanding of changes in metabolite content using a genome-wide association study (GWAS) with eigenvector decomposition (EigenGWAS) on tomato accessions. Results An EigenGWAS was performed on 331 tomato accessions using the first eigenvector generated from the genomic data as a “phenotype” to understand the changes in fruit metabolite content during breeding. Two independent gene sets were identified that affected fruit metabolites during domestication and improvement in consumer-preferred tomatoes. Furthermore, 57 candidate genes related to polyphenol and polyamine biosynthesis were discovered, and a major candidate gene chlorogenate: glucarate caffeoyltransferase (SlCGT) was identified, which affected the quality and diseases resistance of tomato fruit, revealing the domestication mechanism of polyphenols. Conclusions We identified gene sets that contributed to consumer liking during domestication and improvement of tomato. Our study reports novel evidence of selective sweeps and key metabolites controlled by multiple genes, increasing our understanding of the mechanisms of metabolites variation during those processes. It also supports a polygenic selection model for the application of tomato breeding.

Genomic basis of selective breeding from the closest wild relative of large-fruited tomato

“…We used three strategies to identify the selected genomic regions, including EigenGWAS, nucleotide diversity (π) analysis, and XP-CLR methods. For EigenGWAS analysis [ 46 , 47 ], we used the first eigenvector of the Principal Component Analysis (PCA) as “phenotype” and performed GWAS on all tomato accessions using TASSEL software (version 5) [ 48 ] with the default parameters. The top 5% (6.398) windows were determined as candidate selected regions.…”

Editorial: Multi-omics and computational biology in horticultural plants: from genotype to phenotype, volume II

“…Such insights shed light on how genetic variances contribute to the observable characteristics of plants. Yang et al employed an Eigen Genome-Wide Association Study (EigenGWAS) methodology on a collection of 331 tomato varieties to uncover critical genetic determinants that affect metabolite variation throughout the tomato's domestication and enhancement processes, thereby emphasizing the significance of polygenic selection in the developing of tomatoes with superior fruit quality (Yang et al, 2022).…”

Tomato: Breeding and Genomics

“…Kim et al (2021) identified QTL for eight fruit traits in cultivated tomatoes through a genome-wide association study based on 34,550 confident SNPs from the 51 K Axiom® tomato array. Yang et al (2022) investigated a genomewide association study of eigenvectors indicating genetic insights into selective breeding for tomato metabolites in 331 accessions. They identified two independent gene sets of fruit metabolites and 57 candidate genes related to polyphenol and polyamine biosynthesis.…”